Functional impact of anti-PD-L1 treatment on specific lung (tumor) residing myeloid cell subsets

Antibodies targeting Programmed Death-1 (PD-1) or its ligand became a first-line treatment option for metastatic non-small cell lung cancer (NSCLC) patients. Unfortunately, about 75% of patients don’t show any benefit. The main clinically applied biomarker is PD-L1 expression on tumor cells. However, preclinical observations reveal that not PD-L1 on tumor cells, but on tumor infiltrating myeloid cells represents a major determinant for therapy outcome. Further, the lung represents an immunologic organ packed with myeloid subsets that could interact with anti-PD-(L)1 antibody via Fc receptors and/or PD-(L)1. Still the functional impact of anti-PD-L1 therapy on lung myeloid subsets remains largely unknown.
To evaluate the abundance of specific PD-(L)1+ myeloid subsets during NSCLC progression, C57BL/6 mice were intravenously challenged with Lewis Lung Carcinoma (LLC) cells. Next, mice were treated every three days with anti-PD-L1, four times in total. NSCLC engraftment and PD-(L)1 expression was evaluated on weeks 1, 2 and 3 using immunohistochemistry and flow cytometry. Sorted myeloid subsets were functionally evaluated using an ex vivo 3D killing assay, qPCR for alternative checkpoints, arginase1, nitric and super oxide evaluation. Only a marginal therapeutic benefit without increase in CD8+ T cell infiltration was observed upon antiPD-L1 therapy. In contrast, the fraction of Ly6G+ neutrophils decreased during anti-PD-L1 therapy while MHCIIlo F4/80+ macrophages (‘M2’), Ly6C+ inflammatory (IM) and Ly6C- non-classical monocytes (RM) decreased after therapy. Systemically, the ratio of M2/M1 and amount of RMs decreased in bone marrow or/and spleen, respectively. In contrast, PD-L1+ neutrophils and macrophages decreased in blood and spleen while PD-L1+ monocytes only decreased in spleen. Functional evaluation of sorted myeloid subsets showed that in vivo anti-PD-L1 treatment increased: arginase-1 expression in RMs, superoxide production in M2 and CD8+ T-cell stimulating potential of RMs and M1.
These findings could provide novel rationales for potent combination therapies.